Skip Menu |

This queue is for tickets about the Net-SSLeay CPAN distribution.

Report information
The Basics
Id: 118584
Status: rejected
Priority: 0/
Queue: Net-SSLeay
People
Owner: Nobody in particular
Requestors: paul [...] city-fan.org
Cc:
AdminCc:
Bug Information
Severity: (no value)
Broken in: 1.78
Fixed in: (no value)

History Show all quoted text
  Tue Nov 01 07:41:29 2016 paul [...] city-fan.org - Ticket created
Subject: AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Not sure whether this is an AnyEvent or a Net-SSLeay bug, so I though I'd try here first: $ perl -Ilib t/80_ssltest.t 1..415 ok 1 - mode 1 ok 2 - client_connect 1 ok 3 - server_connect 1 Segmentation fault (core dumped) This is the back-trace: #0 internal_find (st=0x300000000, data=data@entry=0x7ffff5ff86f8 <ssl3_ciphers+6552>, ret_val_options=ret_val_options@entry=2) at crypto/stack/stack.c:193 #1 0x00007ffff5ac2b0a in OPENSSL_sk_find (st=<optimized out>, data=data@entry=0x7ffff5ff86f8 <ssl3_ciphers+6552>) at crypto/stack/stack.c:211 #2 0x00007ffff5dccd0e in sk_SSL_CIPHER_find (ptr=0x7ffff5ff86f8 <ssl3_ciphers+6552>, sk=<optimized out>) at include/openssl/ssl.h:826 #3 tls_process_server_hello (s=0xbeafe0, pkt=0x7fffffffdda0) at ssl/statem/statem_clnt.c:1062 #4 0x00007ffff5dcae5f in read_state_machine (s=0xbeafe0) at ssl/statem/statem.c:589 #5 state_machine (s=0xbeafe0, server=0) at ssl/statem/statem.c:385 #6 0x00007ffff5db12da in ssl3_write_bytes (s=0xbeafe0, type=23, buf_=0xf03380, len=2) at ssl/record/rec_layer_s3.c:371 #7 0x00007ffff5dc18d9 in SSL_write (s=<optimized out>, buf=<optimized out>, num=<optimized out>) at ssl/ssl_lib.c:1605 #8 0x00007ffff6045616 in XS_Net__SSLeay_write () from /usr/lib64/perl5/vendor_perl/auto/Net/SSLeay/SSLeay.so #9 0x00007ffff7abf620 in Perl_pp_entersub (my_perl=0x603010) at pp_hot.c:3987 #10 0x00007ffff7ab7ac6 in Perl_runops_standard (my_perl=0x603010) at run.c:41 #11 0x00007ffff7a37b7e in Perl_call_sv (my_perl=<optimized out>, sv=0xbd4ff8, flags=<optimized out>) at perl.c:2807 #12 0x00007ffff4eca1ff in e_cb () from /usr/lib64/perl5/vendor_perl/auto/EV/EV.so #13 0x00007ffff4ec5a06 in ev_invoke_pending () from /usr/lib64/perl5/vendor_perl/auto/EV/EV.so #14 0x00007ffff4edc1cd in ev_run () from /usr/lib64/perl5/vendor_perl/auto/EV/EV.so #15 0x00007ffff4edd31b in XS_EV_run () from /usr/lib64/perl5/vendor_perl/auto/EV/EV.so #16 0x00007ffff7abf620 in Perl_pp_entersub (my_perl=0x603010) at pp_hot.c:3987 #17 0x00007ffff7ab7ac6 in Perl_runops_standard (my_perl=0x603010) at run.c:41 #18 0x00007ffff7a3f816 in S_run_body (oldscope=<optimized out>, my_perl=<optimized out>) at perl.c:2483 #19 perl_run (my_perl=0x603010) at perl.c:2406 #20 0x0000000000400c59 in main (argc=<optimized out>, argv=<optimized out>, env=<optimized out>) at perlmain.c:116 The DEFINE_STACK_OF OpenSSL manual notes: Care should be taken when accessing stacks in multi-threaded environments. Any operation which increases the size of a stack such as sk_TYPE_insert() or sk_push() can "grow" the size of an internal array and cause race conditions if the same stack is accessed in a different thread. Operations such as sk_find() and sk_sort() can also reorder the stack. (forwarded from https://bugzilla.redhat.com/show_bug.cgi?id=1390468)
  Tue Nov 01 20:57:39 2016 mikem [...] airspayce.com - Correspondence added
Subject: Re: [rt.cpan.org #118584] AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Date: Wed, 02 Nov 2016 10:57:23 +1000
To: bug-Net-SSLeay [...] rt.cpan.org
From: Mike McCauley <mikem [...] airspayce.com>
Hi, we saw a similar crash when testing with AnyEvent 7.13, Net-SSLeay 1.78 and openssl-1.1.0-pre5. But we dont see it with the current latest openssl-1.1.0b So, looks like it may be a bug in the prerelease openssl 1.1.0? Cheers. On Tuesday, November 01, 2016 07:41:36 AM paul@city-fan.org via RT wrote: Show quoted text
> Tue Nov 01 07:41:29 2016: Request 118584 was acted upon. > Transaction: Ticket created by paul@city-fan.org > Queue: Net-SSLeay > Subject: AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + > OpenSSL 1.1.0 > Broken in: 1.78 > Severity: (no value) > Owner: Nobody > Requestors: paul@city-fan.org > Status: new > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=118584 > > > > Not sure whether this is an AnyEvent or a Net-SSLeay bug, so I though I'd > try here first: > > $ perl -Ilib t/80_ssltest.t > 1..415 > ok 1 - mode 1 > ok 2 - client_connect 1 > ok 3 - server_connect 1 > Segmentation fault (core dumped) > > This is the back-trace: > > #0 internal_find (st=0x300000000, data=data@entry=0x7ffff5ff86f8 > <ssl3_ciphers+6552>, ret_val_options=ret_val_options@entry=2) at > crypto/stack/stack.c:193 > #1 0x00007ffff5ac2b0a in OPENSSL_sk_find (st=<optimized out>, > data=data@entry=0x7ffff5ff86f8 <ssl3_ciphers+6552>) at > crypto/stack/stack.c:211 #2 0x00007ffff5dccd0e in sk_SSL_CIPHER_find > (ptr=0x7ffff5ff86f8 > <ssl3_ciphers+6552>, sk=<optimized out>) at include/openssl/ssl.h:826 > #3 tls_process_server_hello (s=0xbeafe0, pkt=0x7fffffffdda0) at > ssl/statem/statem_clnt.c:1062 > #4 0x00007ffff5dcae5f in read_state_machine (s=0xbeafe0) at > ssl/statem/statem.c:589 > #5 state_machine (s=0xbeafe0, server=0) at ssl/statem/statem.c:385 > #6 0x00007ffff5db12da in ssl3_write_bytes (s=0xbeafe0, type=23, > buf_=0xf03380, len=2) at ssl/record/rec_layer_s3.c:371 > #7 0x00007ffff5dc18d9 in SSL_write (s=<optimized out>, buf=<optimized out>, > num=<optimized out>) at ssl/ssl_lib.c:1605 > #8 0x00007ffff6045616 in XS_Net__SSLeay_write () from > /usr/lib64/perl5/vendor_perl/auto/Net/SSLeay/SSLeay.so > #9 0x00007ffff7abf620 in Perl_pp_entersub (my_perl=0x603010) at > pp_hot.c:3987 #10 0x00007ffff7ab7ac6 in Perl_runops_standard > (my_perl=0x603010) at run.c:41 #11 0x00007ffff7a37b7e in Perl_call_sv > (my_perl=<optimized out>, sv=0xbd4ff8, flags=<optimized out>) at > perl.c:2807 > #12 0x00007ffff4eca1ff in e_cb () from > /usr/lib64/perl5/vendor_perl/auto/EV/EV.so > #13 0x00007ffff4ec5a06 in ev_invoke_pending () from > /usr/lib64/perl5/vendor_perl/auto/EV/EV.so > #14 0x00007ffff4edc1cd in ev_run () from > /usr/lib64/perl5/vendor_perl/auto/EV/EV.so > #15 0x00007ffff4edd31b in XS_EV_run () from > /usr/lib64/perl5/vendor_perl/auto/EV/EV.so > #16 0x00007ffff7abf620 in Perl_pp_entersub (my_perl=0x603010) at > pp_hot.c:3987 #17 0x00007ffff7ab7ac6 in Perl_runops_standard > (my_perl=0x603010) at run.c:41 #18 0x00007ffff7a3f816 in S_run_body > (oldscope=<optimized out>, > my_perl=<optimized out>) at perl.c:2483 > #19 perl_run (my_perl=0x603010) at perl.c:2406 > #20 0x0000000000400c59 in main (argc=<optimized out>, argv=<optimized out>, > env=<optimized out>) at perlmain.c:116 > > The DEFINE_STACK_OF OpenSSL manual notes: > > Care should be taken when accessing stacks in multi-threaded > environments. Any operation which increases the size of a stack such as > sk_TYPE_insert() or sk_push() can "grow" the size of an internal array and > cause race conditions if the same stack is accessed in a different thread. > Operations such as sk_find() and sk_sort() can also reorder the stack. > > (forwarded from https://bugzilla.redhat.com/show_bug.cgi?id=1390468)
-- Mike McCauley VK4AMM mikem@airspayce.com Airspayce Pty Ltd 9 Bulbul Place Currumbin Waters QLD 4223 Australia http://www.airspayce.com Phone +61 7 5598-7474
  Tue Nov 01 20:57:40 2016 The RT System itself - Status changed from 'new' to 'open'
  Wed Nov 02 04:11:43 2016 paul [...] city-fan.org - Correspondence added
From: paul [...] city-fan.org
On Tue Nov 01 20:57:39 2016, mikem@airspayce.com wrote: Show quoted text
> Hi, > > we saw a similar crash when testing with AnyEvent 7.13, Net-SSLeay > 1.78 and > openssl-1.1.0-pre5. > > But we dont see it with the current latest openssl-1.1.0b > > So, looks like it may be a bug in the prerelease openssl 1.1.0?
Still 100% repeatable with openssl 1.1.0b on Fedora Rawhide: https://kojipkgs.fedoraproject.org//work/tasks/7292/16277292/build.log
  Thu Nov 03 03:10:59 2016 mikem [...] airspayce.com - Correspondence added
Subject: Re: [rt.cpan.org #118584] AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Date: Thu, 03 Nov 2016 17:10:38 +1000
To: bug-Net-SSLeay [...] rt.cpan.org
From: Mike McCauley <mikem [...] airspayce.com>
SIgh: havent been able to get Fedora Rawhide netinst to boot in my VirtualBox VM (Fedora 25 netinst is OK!) So I wont be able to make any progress on this for another 2 weeks as I am travelling. Cheers, On Wednesday, November 02, 2016 04:11:45 AM paul@city-fan.org via RT wrote: Show quoted text
> Queue: Net-SSLeay > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=118584 > > > On Tue Nov 01 20:57:39 2016, mikem@airspayce.com wrote:
> > Hi, > > > > we saw a similar crash when testing with AnyEvent 7.13, Net-SSLeay > > 1.78 and > > openssl-1.1.0-pre5. > > > > But we dont see it with the current latest openssl-1.1.0b > > > > So, looks like it may be a bug in the prerelease openssl 1.1.0?
> > Still 100% repeatable with openssl 1.1.0b on Fedora Rawhide: > https://kojipkgs.fedoraproject.org//work/tasks/7292/16277292/build.log
-- Mike McCauley VK4AMM mikem@airspayce.com Airspayce Pty Ltd 9 Bulbul Place Currumbin Waters QLD 4223 Australia http://www.airspayce.com Phone +61 7 5598-7474
  Tue Nov 22 00:11:32 2016 mikem [...] airspayce.com - Correspondence added
Subject: Re: [rt.cpan.org #118584] AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Date: Tue, 22 Nov 2016 15:11:11 +1000
To: bug-Net-SSLeay [...] rt.cpan.org
From: Mike McCauley <mikem [...] airspayce.com>
Hi again Paul, finally got Fedora Rawhide netinst i386 of 20161031 installed on real hardware :-( again, testing with openssl-1.1.0b Net-SSLeay-1.78 and AnyEvent-7.13 (all compiled locally in a test area) showed no crash in AnyEvent t/80_ssltest.t I saw the same result with Fedora 25. Are you sure your tests have been done with these versions? Cheers. On Thursday, November 03, 2016 05:10:38 PM Mike McCauley wrote: Show quoted text
> SIgh: havent been able to get Fedora Rawhide netinst to boot in my > VirtualBox VM (Fedora 25 netinst is OK!) > > So I wont be able to make any progress on this for another 2 weeks as I am > travelling. > > Cheers, > > On Wednesday, November 02, 2016 04:11:45 AM paul@city-fan.org via RT wrote:
> > Queue: Net-SSLeay > > > > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=118584 > > > > > On Tue Nov 01 20:57:39 2016, mikem@airspayce.com wrote:
> > > Hi, > > > > > > we saw a similar crash when testing with AnyEvent 7.13, Net-SSLeay > > > 1.78 and > > > openssl-1.1.0-pre5. > > > > > > But we dont see it with the current latest openssl-1.1.0b > > > > > > So, looks like it may be a bug in the prerelease openssl 1.1.0?
> > > > Still 100% repeatable with openssl 1.1.0b on Fedora Rawhide: > > https://kojipkgs.fedoraproject.org//work/tasks/7292/16277292/build.log
-- Mike McCauley VK4AMM mikem@airspayce.com Airspayce Pty Ltd 9 Bulbul Place Currumbin Waters QLD 4223 Australia http://www.airspayce.com Phone +61 7 5598-7474
  Tue Nov 22 03:10:58 2016 paul [...] city-fan.org - Correspondence added
From: paul [...] city-fan.org
On Tue Nov 22 00:11:32 2016, mikem@airspayce.com wrote: Show quoted text
> Hi again Paul, > > finally got Fedora Rawhide netinst i386 of 20161031 installed on real > hardware > :-( > > again, testing with > > openssl-1.1.0b > Net-SSLeay-1.78 > and AnyEvent-7.13 > > (all compiled locally in a test area) > showed no crash in AnyEvent t/80_ssltest.t > > I saw the same result with Fedora 25. > > Are you sure your tests have been done with these versions?
Well this is interesting. I've just tried again and an i386 build succeeded but the x86_64 build failed: http://koji.fedoraproject.org/koji/taskinfo?taskID=16560073 If you look at the x86_64 build (http://koji.fedoraproject.org/koji/taskinfo?taskID=16560075), you can see what was installed in the root.log file, and the output from the build in the build.log file. We're now on openssl 1.1.0c but the failure was the same as before.
  Tue Nov 22 16:18:46 2016 mikem [...] airspayce.com - Correspondence added
Subject: Re: [rt.cpan.org #118584] AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Date: Wed, 23 Nov 2016 07:18:28 +1000
To: bug-Net-SSLeay [...] rt.cpan.org
From: Mike McCauley <mikem [...] airspayce.com>
Sigh, Ive been trying to install Rawhide x86_64 20161121 on real hardware but it crashes in anaconda during install. The previously known good rawhide netinst images are no longer available, so I am stumped. Every way I turn to test this bug (if bug is what it is) it disappears before me. I wonder if maybe this is an artifact of your testing environment? Cheers. On Tuesday, November 22, 2016 03:10:59 AM paul@city-fan.org via RT wrote: Show quoted text
> Queue: Net-SSLeay > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=118584 > > > On Tue Nov 22 00:11:32 2016, mikem@airspayce.com wrote:
> > Hi again Paul, > > > > finally got Fedora Rawhide netinst i386 of 20161031 installed on real > > hardware > > > > :-( > > > > again, testing with > > > > openssl-1.1.0b > > Net-SSLeay-1.78 > > and AnyEvent-7.13 > > > > (all compiled locally in a test area) > > showed no crash in AnyEvent t/80_ssltest.t > > > > I saw the same result with Fedora 25. > > > > Are you sure your tests have been done with these versions?
> > Well this is interesting. I've just tried again and an i386 build succeeded > but the x86_64 build failed: > > http://koji.fedoraproject.org/koji/taskinfo?taskID=16560073 > > If you look at the x86_64 build > (http://koji.fedoraproject.org/koji/taskinfo?taskID=16560075), you can see > what was installed in the root.log file, and the output from the build in > the build.log file. We're now on openssl 1.1.0c but the failure was the > same as before.
-- Mike McCauley VK4AMM mikem@airspayce.com Airspayce Pty Ltd 9 Bulbul Place Currumbin Waters QLD 4223 Australia http://www.airspayce.com Phone +61 7 5598-7474
  Wed Nov 23 03:19:58 2016 paul [...] city-fan.org - Correspondence added
From: paul [...] city-fan.org
On Tue Nov 22 16:18:46 2016, mikem@airspayce.com wrote: Show quoted text
> Sigh, > > Ive been trying to install Rawhide x86_64 20161121 on real hardware > but it > crashes in anaconda during install. The previously known good rawhide > netinst > images are no longer available, so I am stumped. > > Every way I turn to test this bug (if bug is what it is) it disappears > before > me.
Perhaps try the Fedora 25 x86_64 installer and then use dnf to upgrade to Rawhide (https://fedoraproject.org/wiki/DNF_system_upgrade)? Show quoted text
> I wonder if maybe this is an artifact of your testing environment?
That's always a possibility but I do my testing by installing into a clean chroot all the dependencies and then do the build there, and using that method I always get the same result. My colleague (who provided the backtrace) was also able to reproduce it. Cheers, Paul.
  Wed Nov 23 22:32:32 2016 mikem [...] airspayce.com - Correspondence added
Subject: Re: [rt.cpan.org #118584] AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Date: Thu, 24 Nov 2016 13:32:17 +1000
To: bug-Net-SSLeay [...] rt.cpan.org
From: Mike McCauley <mikem [...] airspayce.com>
Hi Paul, Sigh, I tried the Point installer to Rawhide option as described on https://fedoraproject.org/wiki/Releases/Rawhide It failed part way through the software download. It failed to download gnome- software, and reported that it tried all the mirrors. Sounds like some sort of problem in the rawhide package repository. There was no option except to exit the installer. Im finding this very frustrating and time consuming for something which may not even be a bug in my software. Can you offer me a reliable way to reproduce this problem? Cheers. On Wednesday, November 23, 2016 03:19:59 AM you wrote: Show quoted text
> Queue: Net-SSLeay > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=118584 > > > On Tue Nov 22 16:18:46 2016, mikem@airspayce.com wrote:
> > Sigh, > > > > Ive been trying to install Rawhide x86_64 20161121 on real hardware > > but it > > crashes in anaconda during install. The previously known good rawhide > > netinst > > images are no longer available, so I am stumped. > > > > Every way I turn to test this bug (if bug is what it is) it disappears > > before > > me.
> > Perhaps try the Fedora 25 x86_64 installer and then use dnf to upgrade to > Rawhide (https://fedoraproject.org/wiki/DNF_system_upgrade)?
> > I wonder if maybe this is an artifact of your testing environment?
> > That's always a possibility but I do my testing by installing into a clean > chroot all the dependencies and then do the build there, and using that > method I always get the same result. My colleague (who provided the > backtrace) was also able to reproduce it. > > Cheers, Paul.
-- Mike McCauley VK4AMM mikem@airspayce.com Airspayce Pty Ltd 9 Bulbul Place Currumbin Waters QLD 4223 Australia http://www.airspayce.com Phone +61 7 5598-7474
  Sat Nov 26 10:49:59 2016 paul [...] city-fan.org - Correspondence added
From: paul [...] city-fan.org
Fix has been applied upstream: http://cvs.schmorp.de/AnyEvent/lib/AnyEvent/Handle.pm?r1=1.248&r2=1.249 Thanks for your help diagnosing this Mike, and sorry about the problems with Rawhide.
  Sat Nov 26 16:24:15 2016 mikem [...] airspayce.com - Correspondence added
Subject: Re: [rt.cpan.org #118584] AnyEvent 7.13 t/80_ssltest.t segfaults with Net-SSLeay 1.78 + OpenSSL 1.1.0
Date: Sun, 27 Nov 2016 07:09:32 +1000
To: bug-Net-SSLeay [...] rt.cpan.org
From: Mike McCauley <mikem [...] airspayce.com>
Hi Paul, On Saturday, November 26, 2016 10:50:00 AM you wrote: Show quoted text
> Queue: Net-SSLeay > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=118584 > > > Fix has been applied upstream: > > http://cvs.schmorp.de/AnyEvent/lib/AnyEvent/Handle.pm?r1=1.248&r2=1.249
Great, but its a shame they didnt see fit to remove the disparaging remarks, and to use the symbolic constants (which were added in 2012) instead of '1|2', Cheers. Show quoted text
> > Thanks for your help diagnosing this Mike, and sorry about the problems with > Rawhide.
-- Mike McCauley VK4AMM mikem@airspayce.com Airspayce Pty Ltd 9 Bulbul Place Currumbin Waters QLD 4223 Australia http://www.airspayce.com Phone +61 7 5598-7474
  Sun Jul 08 12:12:15 2018 chrisn [...] cpan.org - Correspondence added
If I've read this thread correctly, this was caused solely by a bug in AnyEvent which has now been fixed, so I'll mark this as "rejected". Nonetheless, thanks for the report!
  Sun Jul 08 12:12:17 2018 chrisn [...] cpan.org - Status changed from 'open' to 'rejected'